EF = @ We have developeda completelyautomaticalgorithmto quan titativelymeasureleftventricular ejectionfraction(LVEF)from gated @rTc@sestamibi myocardial perfusion SPECT images. Methods: The algorithmoperatesin the three-dimensional space and uses gated short-axis image volumes. It segments the left ventricle(LV), estimates and displaysendocardialand epicardialsurfaces for all gating intervalsin the cardiac cycle, calculatesthe relativeleftventricularcavityvolumesandderives the global EFfrom the end-diastolic and end-systolic volume, all withoutoperatorinteraction.The algorithmfor measuringLVEF was tested in 65 clinicalpatients undergoing16-intervaland 8-interval rest-gated SPECTand validated against &st-pass radionuclideventnculography.Results: Automatic segmenta tion and contouring of the LV was successful in 65/65 (100%) of the studies. Agreement between EFs measured from 8-interval gatedSPECTand EFscalculatedfrom first-passdatawas high (y = 2.44 + 1.03x,r = 0.909,p < 0.001, s.e.e.= 6.87).Agree ment between EF values measured from 16-interval and 8-in tervalgatedSPECTwasexcellent(y = â€”2.7 + 0.97x,r = 0.988, p < 0.001, s.e.e. = 2.65),the latter beingon averagelowerby 3.71percentage points.Conclusion:Ourautomatic methodis rapid and highly agreeswith conventionalradionuclidemea surements ofEF,thusproviding clinically useful additional infor mationto complementmyocardialperfusionstudies.

https://jnm.snmjournals.org/content/jnumed/36/11/2138.full.pdf

Automatic quantification of ejection fraction from gated myocardial perfusion SPECT.

This study investigates the feasibility of routine clinical 201Tl gated perfusion SPECT (gatedTl), and compares quantitative left ventricular ejection fraction (LVEF) and visually-assessed regional wall motion and thickening to analogous values obtained from 99mTc sestamibi gated perfusion SPECT (gated MIBI). Methods: We studied 121 patients with a rest gated 11(3–3.5 mCi, 35 sec/projection)/poststress gated MIBI (25–30 mCi, 25 sec/projection) separate dual-isotope protocol on a 90° dual-detector camera. Automatic quantitation of LVEFs was accomplished using previously developed and validated software, while visual scoring of motion and thickening was performed using four-point scales. Results: Average myocardial counts were lower in gated 11 images (306 ± 81 counts/pixel) compared to gated MIBI images(789 ± 237 counts/pixel). The quality of gated 11 images was ranked as excellent, good, fair and poor in 24.0%, 42.1%, 24.8% and 9.1%, respectively, of the patients, compared to 43.0%, 43.8%, 9.1% and 4.1%, respectively, for gated MIBI images. Quantitative-gated 11 and gated MIBI LVEFs correlated well (y = 0.11+1.05x, r = 0.918, SEE = 6.35). Possible poststress myocardial stunning may have caused gated Tl LVEFs to overestimate gated MIBI LVEFs by a larger (p = 0.03) amount in ischemic patients (n = 47, y = −0.69+1.09x, r = 0.914, s.e.e. = 6.44) compared to nonischemic patients (n = 64, y = −1.58+1.05x, r = 0.919, s.e.e. = 5.93), the residual difference in LVEFs for this latter group being likely due to different isotope resolution in conjunction with small left ventricles. Exact agreement between gated Tl and gated MIBI segmental myocardial function in 41 nonischemic patients was 92.2% (kappa = 0.619) and 95.4% (kappa = 0.586) for motion and thickening scores, respectively. Conclusion: Thallium-201 gated SPECT imaging can be effectively performed on the majority of patients in our clinical environment and offers the opportunity to assess both myocardial perfusion and function using one injection and one imaging sequence, similarly to what is done with 99mTc-based agents.

http://jnm.snmjournals.org/content/38/5/749.full.pdf

Quantitative LVEF and qualitative regional function from gated thallium-201 perfusion SPECT

An investigation of the estimation of ejection fractions and cardiac volumes by a quantitative gated SPECT software package in simulated gated SPECT images

Quantification of gated single-photon emission tomography (SPET) in small hearts has been considered to be inaccurate. To evaluate the validity of gated SPET in a small chamber volume, mathematical simulation and clinical application to paediatric patients were performed. Myocardium with various chamber sizes from 14 ml to 326 ml was generated assuming an arbitrary resolution (6.9-15.7 mm in full-width at half-maximum), noise and zooming factors. The cut-off frequency of the Butterworth filter for preprocessing was varied from 0.16 to 0.63 cycles/cm. The chamber volume was calculated by quantitative gated SPET software (QGS). The patients, aged 2 months to 19 years (n=27), were studied by gated technetium-99m methoxyisobutylisonitrile or tetrofosmin SPET. Image magnification as large as possible was performed during data acquisition to include the whole chest using 1.25-2.0 zooming. Based on the simulation study, an underestimation of the chamber volume occurred below a volume of 100 ml. The degree of underestimation for a 37-ml volume was 49% without zooming, but it improved to 3% with 2x zooming. Filters with a higher cut-off frequency, better system resolution and hardware zooming during acquisition improved quantitative accuracy in small hearts. For the subjects under 7 years old (n=7), quantification of volume and ejection fraction (EF) was possible in 72% of the patients. In those over 7 years old, gated SPET quantification was feasible in all cases. The correlation between gated SPET end-diastolic volume (SPET EDV) and both echocardiographic end-diastolic dimension (EDD) and echocardiographic EDV was good (r=0.84 between SPET EDV and echo EDD, r=0.85 between SPET EDV and echo EDV, P<0.0001 for both). The correlation between gated SPET EF and both echocardiographic fractional shortening (FS) and echocardiographic EF was fair (r=0.69 between SPET EF and echo FS, r=0.72 between SPET EF and echo EF, P<0.0001 for both). In conclusion, quantification of gated SPET of small hearts can be improved by means of a SPET filter with a high cut-off frequency, high system resolution and appropriate zooming. Gated SPET should be attempted not only in patients with small hearts but also in paediatric patients.

/pdf/gated-spet-quantification-of-small-hearts-mathematical-d9eshq6rbq.pdf

Gated SPET quantification of small hearts: mathematical simulation and clinical application.

Assessment of left ventricular diastolic function with electrocardiography-gated myocardial perfusion SPECT: comparison with multigated equilibrium radionuclide angiography.

Myocardial gated tomoscintigraphy with hexakis-(2 methoxy-isobutyl-isonitrile) labelled with 99Tcm, is more suitable to resolve precisely the size of myocardial infarct than nongated 201Tl tomoscintigraphy. Gated tomography gives short axis slices at eight points in the cardiac cycle. A quantitative method to analyse heart wall activity and its motion is proposed. In two groups of patients, one with inferior infarct and the other with anterior infarct, the time-activity curves show a maximum in systole for healthy regions and a flattened curve in akinetic regions. Gated tomoscintigraphy assesses more accurately the size of the injured regions because there is no averaging between systolic and diastolic activity as in 201Tl tomoscintigraphy. This method should permit a better follow-up of patients with myocardial infarct.

T. Oswald

Papers

Myocardial gated tomoscintigraphy with 99Tcm-methoxy-isobutyl-isonitrile (MIBI): regional and temporal activity curve analysis.